This sound nice but after subpoenas start flying many operators will close its doors and the rest will be stored on Amazon aws. Distributed networks are hard tech. Web2 is doomed.
Discussion
Full-nodes could’ve been banned too, but luckily many countries haven’t tried to fight that beast yet. Historically, P2P networks have been attacked a lot more. Instead of whining about pessimistic legal outlooks, try building something to help out the cause.
I’m building consensus here sir
Why do you think IPFS hasn’t been banned yet? It’s legal and quite popular, despite its flaws surrounding node discovery. Illegal content on a few nodes doesn’t make a whole network unviable — it’s those nodes breaking the law, not everyone else who’s apart of the network.
The censorship we see on classical social media often isn’t illegal content, it’s idealogical bias — companies censoring viewpoints they disagree with, despite those views being totally legal to voice.
That’s why I’m here: to stop this unfair power from slipping into the hands of a few giant corporations. If it isn’t illegal and it is media protected by the first amendment, then let it be heard. 🎤
I’ve tried some PoC using WebRTC but I’m stuck at the two-sided-NAT problem. Not having enough time to keep it going due to heavy demand this days.
I haven’t dove into P2P enough yet, been busy getting the Merkle DAG Trees working in a user-server model, but there may be other historic examples we can learn from beyond WebRTC — that also attempt to circumvent the NAT issues.
“2. UPnP and NAT-PMP: Many torrent clients use UPnP (Universal Plug and Play) or NAT-PMP (NAT Port Mapping Protocol) to automatically configure port forwarding on the router. If the router supports these protocols, the torrent client can open a port for incoming connections without manual configuration.
3. Hole Punching: Torrent clients also often use TCP or UDP hole punching. This is similar to the technique used by WebRTC, where two peers each initiate a connection to each other at the same time after having received the external port and address information, which can trick the NAT into thinking that the incoming packets are part of an established connection.”
I was trying the hole-punching method and it worked with only one side-NAT, but two-side made it more complex and I need setup some infra abroad to keep testing. That’s were I stopped because it’s becoming too complex to a non-paid job. And look I don’t even have consensus here. Most people here are happy to be Ethereum 2.0
https://datatracker.ietf.org/doc/id/draft-takeda-symmetric-nat-traversal-00.txt
Just saving it here. Our talk give me some insights.